research article passivation and stabilization of aluminum

Research ArticlePassivation and Stabilization of Aluminum Nanoparticles forEnergetic Materials

Matthew Flannery,1 Tapan G. Desai,1 Themis Matsoukas,2

Saba Lotfizadeh,2 and Matthew A. Oehlschlaeger3

1Advanced Cooling Technologies, Inc., 1046 New Holland Avenue, Lancaster, PA 17601, USA2Penn State University, 150 Fenske Laboratory, University Park, PA 16802, USA3Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY 12180, USA

Correspondence should be addressed to Tapan G. Desai; [email protected]

Received 17 June 2015; Accepted 13 October 2015

Academic Editor: Paulo Cesar Morais

Copyright © 2015 Matthew Flannery et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

In aircraft applications, fuel is used not only as a propellant but also as a coolant and improving both the thermal conductivityand combustion enthalpy of the fuel is beneficial in these applications. These properties can be enhanced by dispersing aluminumnanoparticles into the fuel; however, the nanoparticles require stabilization from agglomeration and passivation from oxidation inorder for these benefits to be realized in aircraft applications. To provide this passivation and stabilization, aluminum nanoparticleswere encapsulated with a coating by the plasma enhanced chemical vapor deposition (PE-CVD) method from toluene precursors.The thermal conductivity, combustion and ignition properties, and stability of the nanoparticles dispersed in RP-2 fuel weresubsequently evaluated. In addition, the effect of dispersing aluminum nanoparticles in RP-2 fuel on the erosion rate of fuel nozzleswas evaluated. The dispersion of PE-CVD coated aluminum nanoparticles at a concentration of 3.0% by volume exhibited a 17.7%and 0.9% increase in thermal conductivity and volumetric enthalpy of combustion, respectively, compared to the baseline RP-2 fuel.Additionally, particle size analysis (PSA) of the PE-CVD coated aluminum nanofuel exhibited retention of particle size over a five-month storage period and erosion testing of a 1mm stainless steel nozzle exhibited a negligible 1% change in discharge coefficientafter 100 hours of testing.

1. Introduction

Increasing the mission capability of aircraft is a never endingobjective for themilitary. Critical to achieving this objective isimproving the thermal performance and energy density of thefuel aboard the aircraft, which is used as both a propellant andcoolant [1]. Thus, improving the thermal conductivity andenergy density of the fuel is one way to reach this objective.The improvement in thermal conductivity of oil-based fluidshas been demonstrated by dispersing metallic nanoparticlesinto the base fluid in numerous research efforts [2]. Addi-tionally, metals have higher combustion energy than carbonbased fuels [3]. Thus, creating a metallic nanofuel for aircraftapplications can improve both the thermal performance and

combustion energy of the fuel to meet the objective of inc-reasing mission capability.

Due to its low density compared to other metals andlarger combustion enthalpy than the base fuel [4], aluminumis a preferred candidate nanoparticle for the development ofnanofuels. The improvement in the volumetric enthalpy ofcombustion of ethanol by Jones et al. shows that additionsof 50 nmdiameter aluminumnanoparticles at concentrationsranging from 1 to 10% by volume exhibited improvementsin the volumetric enthalpy of combustion up to 15%. Thisincrease in heat release upon combustion was also demon-strated by Mitchell et al. for aluminum nanoparticle/dieselfuel systems [5]. Despite the potential for combustion impro-vements, the aluminum rapidly oxidizes to form aluminum

Hindawi Publishing CorporationJournal of NanomaterialsVolume 2015, Article ID 682153, 12 pageshttp://dx.doi.org/10.1155/2015/682153

2 Journal of Nanomaterials

Aluminum nanoparticles

Volatile

Waterbath

precursor

Argon gas

Flow controller

RF generator

Reactor

IPA Magnetic stirrer

LN trap

Pump

(a) (c)

B)

C)CCCC)C)C))C)CCCC)C))

B)B)BBB)B)BB))B)BBBBBBBBBB)B)BB)B)B)B)B)BB)B)B)B)B)BB))B)B)))B)))

≈

Coating

(b)

100nm

Figure 1: PE-CVD coating process wherein metallic nanoparticles are coated by a deeply fragmented volatile organic precursor.

oxide, which reduces the combustion enthalpy of the nano-particles. Additionally, in the dispersed phase, the aluminumnanoparticles can agglomerate due to strong attractive forcesand increase the average particle size of the aluminum, whichcauses the subsequent settling of the nanoparticles. This notonly reduces the stability of the nanofuel, but could increasethe erosion damage of fuel system components.

Reduction from agglomeration can be accomplished bya combination of electrostatic and steric stabilization [6],which can be achieved by the use of surfactants, coatings,or ionic fluids. The surface agents acting to stabilize thenanoparticles from agglomeration simultaneously protectthe nanoparticle from oxidation. Meziani et al. developedstabilized aluminum nanoparticles in organic solvents usinga wet-chemical synthesis approach from alane precursorswith carboxylic acid molecules for surface passivation.Thesemethods demonstrated an average aluminum particle sizeof 48 nm with a 15 nm standard deviation and low oxygenconcentration consistent with passivation from oxidation [7].While the wet-chemical approach has been shown to producesmall diameter nanoparticles that are stabilized and passi-vated during synthesis [7, 8], application of passivating andstabilizing coatings has been conducted in the dry state [9].

To prevent the oxidation and agglomeration of the alu-minum nanoparticles, the PE-CVD coating method wasutilized to develop a 6 nm thick coating that encapsulatesindividual metallic nanoparticles. This coating passivatesthe aluminum nanoparticle to protect from oxidation andreduces the strong attractive forces to weak attractive forces

to eliminate agglomeration, thereby stabilizing the nano-fuel. Through previous research, this coating method hasdemonstrated the successful passivation of aluminum, whichincreased the combustion enthalpy of the aluminum [9].In this research effort, we present the coating of aluminumnanoparticles by the PE-CVD coatingmethod, the evaluationof the stability of the nanofuel over a 5-month period byparticle size analysis (PSA), and the evaluation of the thermalconductivity and combustion enthalpy of the nanofuel atconcentrations of 0.7%, 1.5%, and 3.0% by volume by thetransient hot-wire method and bomb calorimetry, respec-tively. Additionally, the erosion rate of a 1mm stainless steelnozzle with aluminum nanofuel pumped at 200 PSID for 100hours was evaluated in a custom built pumped loop, and theignition delay time of the aluminum nanofuel was measuredby spray injection of the fuel into an electrically heated,constant volume combustion chamber.

2. Materials and Methods

2.1. PE-CVD Coating Deposition. The PE-CVD coating pro-cess is a dry, one-step coating method that applies a nanome-ter thin plasma coating to the surface of the aluminumnanoparticles, which is outlined in Figure 1.

As is shown in Figure 1(a), the PE-CVD process is com-prised of a volatile organic precursor, which is carried byinert argon gas to a plasma reaction chamber that containsaluminum nanoparticles (99.9% purity, 80 nm diameter par-ticles from Nanostructured and Amorphous Materials, Inc.,

Journal of Nanomaterials 3

Houston, TX). The toluene precursor is deeply fragmentedby radio frequency (RF) glow discharge, as depicted inFigure 1(b). The deeply fragmented precursors react on thesurface of the nanoparticles creating a coating thatmimics thechemistry of the initial precursor material [10]. The volatileorganic precursor chosen for this research effort was toluenedue to its chemical similarity to RP-2 fuel. In order to achieveuniform coating, as is shown in Figure 1(c), the aluminumnanoparticles are agitated by a magnetic stirrer. The growthof the coating is linear with respect to time, and therefore, itis controlled by adjusting the residence time of the nanopar-ticles in the reaction chamber.The target coating thickness ofthe toluene PE-CVD coating was 6 nm and was confirmed bytransmission electron microscopy (TEM).

2.2. Nanoparticle Storage Stability Evaluation. To evaluatethe stability of the aluminum nanofuel, the particle sizedistribution of PE-CVD coated aluminum nanofuel samplesat concentrations of 0.7% by volume was evaluated afterinitial dispersion in RP-2 fuel and was reevaluated in one-month intervals for five months. At each month interval, thesamples were sonicated and well mixed, and an aliquot ofthe nanofuel was analyzed by dynamic light scattering (DLS).A baseline nanofuel sample produced from uncoated alu-minum nanoparticles at equivalent concentration was evalu-ated to determinewhether the coating provided stability fromagglomeration during the five-month storage period.

Additionally, to evaluate the ability to redisperse the alu-minum nanoparticles after the first month of storage, theparticle size distribution of an aliquot of PE-CVD coated anduncoated aluminum nanofuel was evaluated with and with-out sonication.

2.3. NanofuelThermal ConductivityMeasurements. The ther-mal conductivity of the aluminum nanofuel was evaluated bythe transient hot-wire method. The transient hot-wire appa-ratus consists of a thin diameter platinum wire, as is shownin Figure 2, immersed in a stagnant fluid sample. Current ispassed through the wire generating heat and the temperatureof the wire is monitored throughout the duration of the testby measuring the electrical resistance of the wire.

The temperature of the wire was determined throughthe resistance-temperature relationship of platinum. Theworking equation of the transient hot-wire test is presentedin (1), where 𝑞 is the applied electrical power to the wire, 𝑘 isthe thermal conductivity of the fluid, Δ𝑇 is the temperaturechange of the wire, 𝑡 is time, 𝛼 is the thermal diffusivity ofthe fluid, 𝑟 is the radius of the wire, and 𝐶 is a constant of theapparatus [11]:

𝑇 (𝑡) − 𝑇ref = Δ𝑇 =𝑞

4𝜋𝑘ln( 4𝛼𝑡

𝑟2𝐶) . (1)

Equation (1) demonstrates that plotting Δ𝑇 versus ln(𝑡)produces a linear plot which can be fitted to determine thethermal conductivity of the fluid. However, prior to analyzingnanofuel samples, the constant of the transient hot-wire test

Instrumentation and power supply connections

Thin platinum wire

Fluid sample holder

Figure 2: Transient hot-wire apparatus for evaluating the thermalconductivity of nanofuels.

apparatus, 𝐶, in (1), was determined by calibrating the appa-ratus with baseline RP-2 fuel using a thermal conductivity of0.15W/mK.

A programmable potentiostat was used to apply powerand record electrical data in the transient hot-wire test appa-ratus, and the data was processed to calculate the thermalconductivity of the nanofuel. The potentiostat was program-med to conduct six consecutive runs in a single trial, whichenables the thermal conductivity to be observed over a five-minute time period. This method not only measured theabsolute thermal conductivity of the nanofuel but also pro-vided indication of the stability of the PE-CVD coated alu-minum nanoparticles.

2.4. Combustion Enthalpy Evaluations. To quantify the inc-rease in combustion energy of the PE-CVD coated aluminumnanofuel, the volumetric enthalpy of combustion was evalu-ated by bomb calorimetry. The evaluations were conductedat concentrations of 0.7%, 1.5%, and 3.0% by volume usinga Parr 1341 oxygen bomb calorimeter. To evaluate the abilityof the coating to passivate the aluminum nanoparticles fromoxidation, the volumetric enthalpy of combustion of baselineuncoated aluminum nanoparticles dispersed in RP-2 fuel atequivalent concentrations was evaluated.

Prior to evaluating the combustion energy of nanofuelsamples, the heat capacity of the bomb calorimeter was deter-mined using National Institute of Standards and Technology(NIST) certified benzoic acid tablets. Additionally, com-bustible gelatin capsules were used to handle the liquid nano-fuel samples and the combustion energy of the capsules wasevaluated prior to conducting combustion energy evaluationson nanofuel samples.

2.5. Nozzle Erosion Testing. Adding aluminum nanoparticlesto RP-2 fuel to increase the thermal conductivity and com-bustion energy presents risks to erosion of fuel system


1

2

34

6

(a)

3 4

5

6

(b)

Figure 3: (a) CADmodel of custom nozzle erosion pumped loop. (b) Cross section of the nozzle and reservoir. (1) Air Driven Piston Pump,(2) Air Pressure Regulator and Flow Meter, (3) High-Accuracy Differential Pressure Sensor, (4) Adjustable Pressure Relief Valve, (5) 1mmstainless steel nozzle, and (6) Nanofuel Reservoir.

components. The most vulnerable area was suspected tobe the injectors where the nanofuel is forced through asmall diameter orifice and atomized. Nanofluids that areunstable will agglomerate during storage and when pumpedthrough the nozzle could severely erode the nozzle or evenclog the nozzle. To evaluate the effect of the addition ofPE-CVD coated aluminum particles to RP-2 fuel, a custom-made pumped loop was designed and fabricated to pumpthe PE-CVD coated nanofuel at a concentration of 0.7% byvolume through a 1mm stainless steel nozzle at 200 PSID for100 hours, as shown in Figure 3.

The nozzle erosion pumped loop was designed withan Air Driven Piston Pump that delivered nanofuel to aremovable 1mm nozzle. A differential pressure transducerwas connected to the inlet and outlet of the nozzle tomeasureand monitor the differential pressure across the nozzle. Apressure relief valve was incorporated into the pumped loopto ensure fluid bypass in the event of clogging.

Erosion of the nozzle would cause a change in the flowcharacteristics of the nozzle. Thus, a nozzle evaluation loopwas developed, as shown in Figure 4, to measure the dis-charge coefficient of the nozzle at 25-hour intervals. Addi-tionally, optical microscopy evaluations were conducted tomeasure the diameter of the nozzle and observe the nozzlefor signs of erosion.

At 25-hour intervals, the nozzle was removed from theerosion test pumped loop, cleaned thoroughly in acetone, andplaced into the nozzle evaluation loop. Water was pumpedthrough the nozzle at flow rates varying from 2.5GPH to15GPH in 2.5GPH increment and the pressure differentialacross the nozzle at steady state was measured. The flowperformance and erosion of the nozzle were compared toa baseline RP-2 fuel tested for 100 hours under the same

Gear pump

Pressure transducers

Nozzle

Flow meter

Figure 4: Nozzle evaluation test loop developed to characterize thedischarge coefficient of the nozzle at 25-hour intervals.

condition in order to determine how the addition of metallicnanoparticles effects the erosion of fuel system components.

2.6. Ignition Delay TimeMeasurements. To evaluate the effectof the addition of the aluminum nanoparticles on the com-bustion kinetics of the fuel, the ignition delay time of baselineRP-2 fuel, uncoated aluminumnanofuel, andPE-CVDcoatedaluminum nanofuel, each at a concentration of 1.5% by vol-ume, was evaluated in a constant volume, electrically heatedcombustion chamber, shown in Figure 5.


Accumulator

Solenoid valve

Injector

Insulated and heatedspray chamber

Piezoelectric pressure

Exhaust

High-pressurefuel/nanofuel

Filltransducer (to DAQ)

d = 10 cm, l = 12 cm

Figure 5: Spray combustion apparatus for evaluating the ignition delay time of RP-2 fuel and aluminum nanofuel.

The combustion chamber was heated with electric-resist-ance heaters embedded into the chamber walls and was insu-lated with 5 cm of high-temperature ceramic fiber, mineralwool insulation. The chamber was instrumented with K-type thermocouples for chamber wall temperature measure-ments and a water-cooled Kistler pressure transducer (model6041A) for measurement of chamber pressure during thespray ignition event and determination of the ignition delaytime. At the top of the cylindrical combustion chamber, alow-pressure diesel, single-hole, pintle-style injector (BoschW0133-1827210) atomized the fuel samples into the chamber.The injector was cooled by circulating diesel fuel, which flowsthrough the injector body. The fuel sample was pressurizedin a hydraulic accumulator, and the duration of injectionwas controlled by a solenoid valve located between the high-pressure accumulator and the injector. The pressure of thefuel delivered to the injector was set to 2200 psia for a dura-tion of 3ms.

Prior to ignition delay time measurements, the massof the fuel injected was characterized to ensure consistentinjector performance. High-pressure fuel was injected intoa container and the mass of single injections was measuredover ten trials using a high-accuracy scale (Mettler PM6100).The mass per injection for the three fuel samples studied ispresented in Figure 6.

As is shown in Figure 6, the mass of fuel injected with a2200 psia injection pressure and a 3ms pulse duration wasconsistent over the 10 trials and deviated by less than 3%between fuel samples. Additionally, the consistent injectionperformance between the baseline RP-2 fuel and the nanofuelsamples demonstrated that injector performance was notaffected by the addition of toluene PE-CVDcoated aluminumnanoparticles.

To carry out a series of spray ignition delay timemeasure-ments, the combustion chamber was heated to a temperatureof approximately 800K, and the heaters were turned off. Asthe combustion chamber cooled over the course of several

0

20

40

60

80

100

120

140

Mas

s inj

ecte

d pe

r inj

ectio

n (m

g)

0 2Injection number

4 6 8 10

1.5% uncoated nAl/RP-2, avg. = 109mg, st. dev.= 9mg1.5% coated nAl/RP-2, avg. = 115mg, st. dev. = 15mgRP-2, avg. = 116mg, st. dev. = 13mg

Figure 6: Characterization of fuel mass injection during spray com-bustion testing demonstrating consistent spray performance.

hours, spray injection measurements were conducted attemperatures ranging from 800 to 600K with a temperatureuniformity of ±3K. The evacuated chamber was filled withhigh-purity air to 20 atm from a compressed gas cylinder,and the fuel sample was injected into the high-temperature,high-pressure air, which initiates the combustion experiment.The pressure of the chamber was monitored using a Kistlerpressure transducer, as shown in Figure 7.

As is shown in Figure 7, at injection, the pressure in thechamber immediately decreases due to evaporative coolingof the chamber air as the fuel evaporates in the chamber andmixes. After an induction period, a partially mixed fuel/air


0.20

0.15

0.10

0.05

0.00

Ignition

Injection

Energyrelease

0 5 201510Time (ms)

Pres

sure

sign

al (V

)

−0.05

−10 −5

Ignition delay = 11.1msInjected into 20atm air at 698K1.5% coated nAl/RP-2

Evaporative cooling

Figure 7: Sample pressure profile for spray ignition delay measure-ments.

region locally ignites causing the chamber pressure to riserapidly. The ignition delay time was determined by compar-ing the time interval between spray injection, defined as thetime at which the chamber pressure first decreases due tospray evaporation, and the onset of ignition, defined as thetime at which the pressure gradient first becomes positive dueto ignition-related heat release.

3. Results and Discussion

3.1. PE-CVD Coating Deposition. The target coating thick-ness of the PE-CVD coating from toluene precursor was6 nm. To ensure the coating thickness met this target andencapsulated individual nanoparticles, the coating thicknessof dry aluminum nanoparticles was characterized by trans-mission electronmicroscopy (TEM). As is shown in Figure 8,the toluene PE-CVD coating encapsulates the individualaluminum nanoparticles and has an average thickness of6.21 nm, which demonstrates that PE-CVD coating met thetarget coating thickness specification.

3.2. Nanoparticle Stability Evaluation. During storage, nano-particles will settle due to the relative density of the particleand the base fuel. However, a stabilized nanofluid will redis-perse after settling without increasing particle size (i.e., willnot agglomerate). The ability to redisperse the PE-CVDcoated aluminum nanofuels compared to the uncoated coun-terpart was evaluated by measuring the median particle sizewith andwithout sonication of the respective sample by dyna-mic light scattering (DLS) after onemonth of storage at roomtemperature.

As is shown in Figure 9, the PE-CVD coated nanofuelsample increases particle size over the one-month periodbut is easily redispersed with sonication. This indicates thatweak bonding forces are present between agglomerated alu-minum nanoparticles. Alternatively, the uncoated aluminumnanofuel sample demonstrated a consistent particle size that

7.246nm5.391nm

5.982nm

Figure 8: TEM micrograph of toluene PE-CVD coated aluminumnanoparticles exhibiting a 6 nm coating thickness that encapsulatesindividual aluminum nanoparticles.

Weak bonding forces broken by

agitation

PECVD coated nanofuel sampleUncoated nanofuel sample

Irreversible agglomeration

Initial withsonication

1-month storagewithout sonication

1-month storagewith sonication

050

100150200250300350400450500

Med

ian

part

icle

size

(nm

)

Figure 9: Median particle size of PE-CVD coated and uncoatedaluminum nanofuel after one month of storage before and aftersonication demonstrating that the PE-CVDcoated nanoparticles areable to be redispersed.

could not be redispersed due to strong bonding forcesbetween aluminum nanoparticles which leads to agglomera-tion. The ability to redisperse the PE-CVD coated aluminumnanofuel sample by sonication demonstrates that the coatingprovides stabilization from agglomeration, which improvesstorage stability of the nanofuel.

In order to quantify the extent of agglomeration of PE-CVDcoated anduncoated aluminumnanoparticles, dynamiclight scattering (DLS) was conducted to evaluate the particlesize distribution of each sample. Particle size distributionswere determined by the number of particles of a given diam-eter. Moreover, the samples were stored at room temperatureand the DLS analysis was repeated at one-month intervalsover a 5-month period to determine the growth rate of parti-cle size due to agglomeration. DLS results at the one-monthintervals are presented in Figure 10.


101

102

103

104

Particle size (nm)

0

5

10

15

20

25

30

35

40

Mea

n nu

mbe

r (%

)

Uncoated t = 2 monthsPECVD coated = 5 months

PECVD coated t = 2 months

Uncoated t = 1 monthPECVD coated t = 1 monthUncoated t = 0 monthsPECVD coated t = 0 months

Figure 10: Dynamic light scattering results of PE-CVD coated anduncoated aluminum nanoparticles dispersed in RP-2 fuel demon-strating that the PE-CVD coated nanofuel maintained a reducedmedian particle size over a five-month storage period.

As is shown in Figure 10, the smallest nanoparticle sizepresent in the uncoated sample distribution was greater than200 nm, while the PE-CVD coated aluminum nanoparticledistribution exhibited the smallest aluminum nanoparticlesize at ∼80 nm. Moreover, the median particle size of thePE-CVD coated aluminum nanoparticle sizes was ∼150 nm,which was 30% of the median particle size of the uncoatedcounterpart. After being stored for five months at roomtemperature, the PE-CVD coated aluminum nanoparticlesmaintained a smaller particle size than the uncoated counter-part sample with 90% of the particles being less than 235 nmin diameter indicating that the PE-CVD coating preventsagglomeration of the aluminum nanoparticles.

3.3. NanofuelThermal ConductivityMeasurements. The ther-mal conductivity of PE-CVD coated aluminum nanofuelsamples at a concentration of 0.7%, 1.5%, and 3.0% byvolume was determined by the transient hot-wire methodin nine trials spanning a three-month period and results arepresented in Figure 11. The measured thermal conductivitywas plotted with the upper and lower bounds of the Hashinand Shtrikman (H-S) model for thermal conductivity innanofluids, which is presented in (2), where 𝑘f is the thermalconductivity of the base fluid RP-2 fuel, 𝑘p is the thermalconductivity of the aluminum particle, 𝜙 is the volumefraction of nanoparticles, [𝑘] = 𝑘p −𝑘f , and 𝑘nf is the thermalconductivity of the nanofuel [12]:

𝑘f [1 +3𝜙 [𝑘]

3𝑘f + (1 − 𝜙) [𝑘]] ≤ 𝑘nf

≤ [1 −3 (1 − 𝜙) [𝑘]

3𝑘p − 𝜙 [𝑘]] 𝑘p.

(2)

Ther

mal

cond

uctiv

ity (W

/mK)

0.150.160.170.180.19

0.20.210.220.230.240.250.26

0.5 1 1.5 2 2.5 3 3.50Volume fraction (%)

Experimental dataH-S upper bound

H-S lower bound

Figure 11: Thermal conductivity of PE-CVD coated aluminumnanoparticles dispersed in RP-2 fuel at 0.7%, 1.5%, and 3.0% volumefractions.

As is shown in Figure 11, the thermal conductivity of PE-CVD coated aluminum nanofuel increases with increasingnanoparticle concentration and reaches a 17.7% improvementin thermal conductivity over the baseline RP-2 fuel at a 3.0%volume fraction. The error bars in Figure 11 represent onestandard deviation in the thermal conductivity measuredover multiple trials.This indicates that addition of aluminumnanoparticles improves the thermal performance of RP-2 fueland, as a result, increases the heat transfer properties of thefuel as a coolant. Furthermore, agreement of the thermal con-ductivity of the PE-CVD coated aluminum nanofuel sampleswith the lower bound of the H-S model indicates that thereis no anomalous improvement in thermal performance. Thewell-known H-S model bounds the thermal conductivity ofa nanofluid based on the configuration of the nanoparticles.The lower bound of the model corresponds to a nanofluidconfiguration in which the nanoparticles are discretely dis-persed, while the upper bound of the model correspondsto a nanofluid configuration in which nanoparticles formchains that increase the thermal conduction path, as shownin Figure 12 [12]. The agreement with the lower bound of theH-S model provides supporting evidence that the PE-CVDcoated aluminum nanofuels have increased stability that canbe attributed to lack of agglomeration and support the PSAresults presented in Figure 10.

As nanoparticles settle, the thermal conductivity of thenanofuel will decrease, approaching that of the RP-2 fuel.Hence, a consistent thermal conductivity throughout the sixconsecutive runs indicates a low settling rate and demon-strates a stable nanoparticle suspension. The experimentallydetermined thermal conductivity of six consecutive runs forthe RP-2 baseline, 0.7%, 1.5%, and 3.0% by volume PE-CVDcoated aluminum nanofuel is presented in Figure 13.

As is shown in Figure 13, the thermal conductivity ofthe nanofuel samples is consistent over six consecutive runs,


Liquid medium

(a)

Liquid medium

(b)

Figure 12: Depiction of nanoparticle configurations that correspond to the (a) lower bounds and (b) upper bounds of the H-S model [12].

RP-2 0.71.5 3.0

1.5 2 2.5 3 3.5 4 4.5 5 5.5 61Run number

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0.26

Ther

mal

cond

uctiv

ity (W

/mK)

Figure 13: Thermal conductivity measurements of baseline RP-2,0.7%, 1.5%, and 3.0% PE-CVD coated aluminum nanofuel over sixconsecutive runs spanning a five-minute period exhibiting consis-tent thermal conductivity and indicating low settling rate of thealuminum PE-CVD nanoparticles.

which indicates a low settling rate. The consistent thermalconductivity between consecutive runs provides evidence tosupport enhanced stabilization of the PE-CVD coated alu-minum nanoparticles.

PE-CVD coated aluminum nanoparticle stability in stor-age conditions was also determined by repeating the ther-mal conductivity analysis over three consecutive months.Agglomerating nanoparticleswill increase the diameter of thenanoparticles causing them to settle out of solution, whichreduces the thermal conductivity of the nanofuel, accordingto the H-S model. Consistent thermal conductivity readings

0.15

0.16

0.17

0.18

0.19

0.2Th

erm

al co

nduc

tivity

(W/m

K)

2 31Month

Pure RP-20.7% PECVD coated nanofuel1.5% PECVD coated nanofuel3.0% PECVD coated nanofuel

Figure 14: Thermal conductivity of baseline RP-2, 0.7%, 1.5%, and3.0% PE-CVD coated aluminum nanofuel after three months ofstorage indicating the ability to easily redisperse the nanoparticlesafter storage.

over a three-month period would demonstrate a stable nano-particle suspension that resists agglomeration and subse-quent settling.Thermal conductivity data conducted over thethree-month period is presented in Figure 14.

As is shown in Figure 14, the thermal conductivity of thebaseline RP-2, 1.5%, and 3.0% samples was repeatable over thethree-month period and exhibited a 2.6% relative standarddeviation at 3.0% volume fraction. The repeatable thermalconductivity over a three-month storage period demonstra-ted a nanoparticle suspension that was easily redispersed dueto stabilizing the aluminum nanoparticles from agglomera-tion and supports the data presented in Figure 9.


3.4. Combustion Enthalpy Evaluation. Thevolumetric enthal-py of combustion,Δ𝐻fuel,vol, of PE-CVDcoated and uncoatedaluminum nanofuel samples at a concentration of 0.7%, 1.5%,and 3.0% by volume was determined by bomb calorimetryand was calculated from (3), where 𝐶cal is the heat capacityof the bomb calorimeter, Δ𝑇 is the temperature change ofthe bomb calorimeter, 𝑚gel is the mass of the gelatin capsuleused to hold the liquid sample, Δ𝑈gel is the internal energyof combustion of the gelatin capsule, 𝐿wire is the length ofignition wire combusted during the bomb calorimetry trials,𝛾wire is the combustion energy of the wire per unit length, Δ𝑛is the change in the number of moles of gaseous species, 𝑅 isthe universal gas constant (8.3145 J/molK), 𝜌fuel is the densityof the fuel sample, and𝑚fuel is the mass of the fuel sample:

Δ𝐻fuel,vol

=(𝐶calΔ𝑇 − 𝑚gelΔ𝑈gel − 𝐿wire𝛾wire + Δ𝑛𝑅𝑇) 𝜌fuel

𝑚fuel.

(3)

Experimentally determined volumetric combustion enthal-pies of PE-CVD coated and uncoated aluminum nanofuelsamples that were stored for one month are presented inFigure 15.

As shown in Figure 15, the volumetric enthalpy of com-bustion of the PE-CVD coated aluminum nanofuels inc-reased with increasing volume concentration, reaching 0.9%enhancement at 3.0% volume fraction. Conversely, uncoated0.7% and 1.5% volume fractions did not improve the volumet-ric combustion enthalpy of the RP-2 fuel, while the uncoated3.0% nanofuel sample exhibited 0.3% improvement in volu-metric combustion enthalpy. Furthermore, statistical analysisof the combustion energy data using a 𝑡-test, assumingunequal variance, suggests that the measured improvementof the 3.0% volume fraction PE-CVD coated sample is withina 90% confidence interval. Thus, the statistical analysis ofthe volumetric enthalpy of combustion suggests that theobserved improvement in energy density of the PE-CVDcoated samples was statistically significant.

3.5. Nozzle Erosion Testing. After 25 hours of erosion testing,the nozzle was placed in the evaluation test loop and waterwas pumped through the nozzle at flow rates varying from 2.5to 15GPH in 2.5GPH increments. The pressure differentialacross the nozzle was measured in three trials. The dischargecoefficient, 𝐶

𝑑, of the nozzle was determined by (4), where

Δ𝑃theor is the theoretical pressure difference of the nozzle atthe given flow rate and Δ𝑃meas was the measured pressuredifference. The theoretical pressure difference was calculatedfrom (5), where �� is the mass flow rate of water, 𝐷 is thediameter of the nozzle (1mm), 𝜌 is the density of water, and𝑑 is the ratio of the nozzle diameter to the upstream pipediameter (0.16):

𝐶𝑑=√Δ𝑃theor

√Δ𝑃meas(4)

Δ𝑃theor = (4��

𝜋𝐷2𝜌)

2 𝜌 (1 − 𝑑4)

2. (5)

37300373503740037450

3755037600376503770037750378003785037900

38000

Volu

met

ric en

thal

py o

f com

busti

on (M

J/m3)

0.5 1 1.5 2 2.5 3 3.5 40Aluminum nanoparticle volume fraction (%)

PECVD coated nanofuelUncoated nanofuel

37950

37500

Figure 15: Volumetric combustion enthalpy of PE-CVD coated anduncoated aluminum nanofuel samples.

NanofuelBaseline RP-2

10 20 30 40 50 60 70 80 90 100 110 1200Erosion time (hours)

0.8

0.81

0.82

0.83

0.84

0.85

0.86

0.87

0.88

0.89

0.9

Disc

harg

e coe

ffici

ent

Figure 16: Nozzle discharge coefficients throughout the 100 hours oferosion testing indicating negligible change in discharge coefficient.

The discharge coefficients of the nozzle tested with baselineRP-2 fuel and 0.5% volume fraction nanofuel throughout the100-hour erosion test are presented in Figure 16.

As is shown in Figure 16, after 100 hours of erosion test-ing, the discharge coefficient of the nozzle tested with nano-fuel reduced by 2% while the baseline RP-2 fuel decreasedby 1% indicating a negligible change in discharge coefficientfor both samples. The error bars in Figure 16 represent onestandard deviation in themeasured discharge coefficient overmultiple experimental trials. Optical microscopy images of


Nozzle discharge, nanofuel (t = 100 hours) Nozzle discharge, RP-2 fuel (t = 100 hours)

Nozzle inlet, RP-2 fuel (t = 100 hours)Nozzle inlet (t = 0 hours)

Nozzle discharge (t = 0 hours)

Nozzle inlet, nanofuel (t = 100 hours)

Figure 17: Optical microscopy images of nozzle discharge and inlets at initial conditions and after 100 hours of erosion testing exhibiting nosignificant erosion damage.

the nozzle discharge and inlet, prior to testing and after 100hours of erosion testing, are presented in Figure 17.

As is shown in Figure 17, the diameter of the nozzle dis-charge and inlet remained constant after 100 hours of erosiontesting. However, additional wear was noticed in the inlet ofthe nanofuel tested nozzle after 100 hours of erosion testing.The additional wear in the nozzle inlet was suspected toincrease the roughness of the inlet, thereby increasing thecoefficient of friction and subsequently reducing the dis-charge coefficient, as is observed in Figure 16. However, thechange in discharge coefficient between the nanofuel testednozzle and baseline RP-2 tested nozzle was insignificant after100 hours of pumping the fuel at a pressure differential of200 PSI indicating that the nanofuel did not increase theerosion rate of fuel injector components.

3.6. Ignition Delay TimeMeasurements. To evaluate the effectof adding aluminum nanoparticles to the combustion kinet-ics of RP-2 fuel, spray ignition delay times were measured ina constant volume combustion chamber. Spray ignition delay

times at temperatures ranging from 600 to 800K for baselineRP-2 fuel, 1.5%PE-CVDcoated aluminumnanofuel, and 1.5%uncoated aluminum nanofuel are presented in Figure 18.

The spray ignition delay time for the PE-CVD coatedaluminum nanofuels is 10–20% longer than baseline RP-2fuel, which is considered within an acceptable deviation. Theignition delay times of the uncoated RP-2 nanofuel sampleswere approximately 50–100% longer than the baseline RP-2. The combustion process is divided into three major steps:evaporation, mixing, and gas-phase chemical reaction. It washypothesized that the uncoated sample formed agglomeratesthat reduced the evaporation rate of the RP-2 fuel, whichreduces the kinetics of the combustion process. On the otherhand, the negligible change in ignition delay time for the PE-CVD coated aluminum nanofuel sample provides evidenceto support the conclusion that the PE-CVD coating reducesagglomeration and permits the fuel to evaporate and main-tain combustion kinetic characteristics while simultaneouslyreleasing more energy as determined in the combustionenthalpy evaluations.


1

10

100

Spra

y ig

nitio

n de

lay ti

me (

ms)

1.30 1.35 1.40 1.45 1.50 1.55 1.60 1.651.251000/T (1/K)

800 780 760 720740 700 680 660 640 620Temperature (K)

Fuel spray injected into 20atm heated airRP-21.5% coated nAl/RP-21.5% uncoated nAl/RP-2

Figure 18: Spray ignition delay time results demonstrating a 10–20%increase in ignition delay time for the PE-CVD coated aluminumnanofuel and a 50–100% increase in ignition delay time for theuncoated counterpart.

4. Conclusions

Utilizing nanoparticle technology to improve the thermal per-formance and energy density of fuels can helpmeet the objec-tives of improving mission capabilities of aircraft but must bepassivated from oxidation and stabilized from agglomerationto ensure the functionality of the fuel is not compromised.Through our research, a passivating 6 nm thick coating thatstabilizes individual nanoparticles was demonstrated by thePE-CVD coating process. Once dispersed into RP-2 base fuel,the nanoparticles exhibited retention of particle size over afive-month storage period, thereby demonstrating stabilityfrom agglomeration. Additionally, the thermal conductivityand volumetric enthalpy of combustion of the PE-CVDcoated aluminum nanofuel at a concentration of 3.0% byvolume exhibited 17.7% and 0.9% improvement, respectively,compared to the baseline RP-2 fuel. At an equivalent concen-tration, uncoated aluminum nanofuel exhibited 0.3% imp-rovement in volumetric enthalpy of combustion compared tothe baseline RP-2 fuel. The greater improvement in the volu-metric enthalpy of combustion between the PE-CVD coatedand uncoated aluminumnanofuel indicates that the PE-CVDcoating protected the aluminum fromoxidation, thereby pro-viding a greater amount of combustible aluminum per massof nanofuel. Furthermore, the PE-CVD coated aluminumnanofuel exhibited a negligible 1% reduction in the dischargecoefficient of a 1mm stainless steel nozzle compared to base-line RP-2 fuel after 100 hours of erosion testing at pressuredifferential across the nozzle of 200 PSID. Finally, the ignitiondelay time of the PE-CVD coated aluminum nanofuel and

uncoated aluminum nanofuel at a concentration of 1.5% byvolume exhibited an increase in ignition delay time of 10–20% and 50–100%, respectively, compared to the baseline RP-2 fuel. Thus, the larger increase in ignition delay time for theuncoated sample compared to the PE-CVD coated sampleindicates that the agglomeration of the uncoated nanoparti-cles affected the ignition process. As a result, the passivationand stabilization of aluminum nanoparticles by the PE-CVDcoating method demonstrated that nanoparticle technologycan be utilized to improve the thermal performance andenergy density of fuels.

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper.

Acknowledgments

The authors thank the United States Air Force for theirsupport of these developments under Contract no. FA9300-13-M-1007. Additional support was provided by the NationalScience Foundation under Grant no. CBET GOALI no.1132220.

References

[1] P. De Bock and B. Gerstler, Thermal Challenges in Today’sCommercial and Military Aircraft, Electronics Cooling, 2011.

[2] R. A. Bhogare and B. S. Kothawale, “A review on applicationsand challenges of nano-fluids as coolant in automobile radiator,”International Journal of Scientific and Research Publications, vol.3, no. 8, 2013.

[3] Y. Gan and L.Qiao, “Combustion characteristics of fuel dropletswith addition of nano and micron-sized aluminum particles,”Combustion and Flame, vol. 158, no. 2, pp. 354–368, 2011.

[4] M. Jones, C. H. Li, A. Afjeh, and G. Peterson, “Experimentalstudy of combustion characteristics of nanoscale metal andmetal oxide additives in biofuel (ethanol),” Nanoscale ResearchLetters, vol. 6, no. 1, article 246, 2011.

[5] M. R. Mitchell, R. E. Link, M.-J. Kao, C.-C. Ting, B.-F. Lin,and T.-T. Tsung, “Aqueous aluminum nanofluid combustion indiesel fuel,” Journal of Testing and Evaluation, vol. 36, no. 2,Article ID 100579, 2008.

[6] A. Kraynov and T. E. Muller, Concepts for the Stabilization ofMetal Nanoparticles in Ionic Liquids, InTech Open Access, 2011.

[7] M. J. Meziani, C. E. Bunker, F. Lu et al., “Formation andproperties of stabilized aluminum nanoparticles,” ACS AppliedMaterials & Interfaces, vol. 1, no. 3, pp. 703–709, 2009.

[8] R. J. Helmich andK. S. Suslick, “Chemical aerosol flow synthesisof hollow metallic aluminum particles,” Chemistry of Materials,vol. 22, no. 17, pp. 4835–4837, 2010.

[9] A. Shahravan, T. Desai, and T. Matsoukas, “Passivation ofaluminum nanoparticles by plasma-enhanced chemical vapordeposition for energetic nanomaterials,” ACS Applied Materials& Interfaces, vol. 6, no. 10, pp. 7942–7947, 2014.

[10] A. Shahravan, T. Desai, and T.Matsoukas, “Controlledmanipu-lation of wetting characteristics of nanoparticles with dry-basedplasma polymerization method,” Applied Physics Letters, vol.101, no. 25, Article ID 251603, 2012.


[11] H. M. Roder, “A transient hot wire thermal conductivity appa-ratus for fluids,” Journal of Research of the National Bureau ofStandards, vol. 86, no. 5, pp. 457–493, 1981.

[12] J. Eapen, R. Rusconi, R. Piazza, and S. Yip, “The classical natureof thermal conduction in nanofluids,” Journal of Heat Transfer,vol. 132, no. 10, Article ID 102402, 2010.

Submit your manuscripts athttp://www.hindawi.com

ScientificaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

CorrosionInternational Journal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Polymer ScienceInternational Journal of



CeramicsJournal of


CompositesJournal of

NanoparticlesJournal of



International Journal of

Biomaterials


NanoscienceJournal of

TextilesHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Journal of

NanotechnologyHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

CrystallographyJournal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


CoatingsJournal of

Advances in

Materials Science and EngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Smart Materials Research



MetallurgyJournal of


BioMed Research International

MaterialsJournal of


Nano

materials


Journal ofNanomaterials

research article passivation and stabilization of aluminum

Documents